Wednesday, February 17, 2010

The rise of globally synthesized collective intelligence

In the first chapter of Building Web Reputation Systems, Randy Farmer uses one of those priceless quotes from Yogi Berra;
“Nobody goes there anymore—it’s too crowded.”
We can guess what it means -- it's a funny turn of a phrase, but it makes sense, i.e. the place that Yogi was referring to had been so obvious to the masses that it was so popular, it had become overcrowded. When he said this, you can bet that Yogi and his friends [who weren't "nobodies" who followed crowds] had already moved on to the next big trendy thing ... when you're hip, you don't go where the crowds are.  In the 1950s reputations mattered more because they were still primarily changed by word of mouth -- television was still in its early days and there was nothing like the internet for increasing the speed of news and communication ... it was easier to be hip, cool, one step ahead of the crowd. Reputation still matters because it is information used to make a value judgment about an object or a person -- how that information is transferred from one person to another has changed dramatically, but many of the principles are the same.

Increasingly we rely upon instantaneously updated hyper-current news from things like Twitter along with globally synthesized collective intelligence to furnish with that reputation information -- we Google something, we check Yelp!, we have our different online sources ... if the decision is serious, we spend a fair amount of time looking -- but the tools we use have gotten a lot more powerful; we have also become more adept at using searching engines and determining the trustworthiness of difference sources.  For some time, we have been familiar with a vast array of reputation systems like FICO scores, Nielsen ratings, Google PageRank ...we might recognize that none of these are perfect, but we still use them to make inferences ... to sort wheat from chaff.

The problem of course, is that we can't control how people will use the reputation systems ... we can avoid doing something stupid like Fair Issac did in marketing FICO scores as a way to tell whether a job candidate is going to be a good employee ... but it is very tricky about how reputations are built ... that is why there is going to be an even bigger market for reputation defenders.